US5157481A - Registration and contour correction circuit and method for solid-state camera - Google Patents

Registration and contour correction circuit and method for solid-state camera Download PDF

Info

Publication number
US5157481A
US5157481A US07/810,360 US81036091A US5157481A US 5157481 A US5157481 A US 5157481A US 81036091 A US81036091 A US 81036091A US 5157481 A US5157481 A US 5157481A
Authority
US
United States
Prior art keywords
data
registration
interpolation
solid
correction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
US07/810,360
Other languages
English (en)
Inventor
Naoki Kashimura
Kazuhiro Ban
Jun Hattori
Yoshiro Aoyagi
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ikegami Tsushinki Co Ltd
Original Assignee
Ikegami Tsushinki Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP1290098A external-priority patent/JP2552741B2/ja
Priority claimed from JP1290099A external-priority patent/JP2552742B2/ja
Application filed by Ikegami Tsushinki Co Ltd filed Critical Ikegami Tsushinki Co Ltd
Application granted granted Critical
Publication of US5157481A publication Critical patent/US5157481A/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • H04N9/646—Circuits for processing colour signals for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/10—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from different wavelengths
    • H04N23/13—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from different wavelengths with multiple sensors
    • H04N23/15—Image signal generation with circuitry for avoiding or correcting image misregistration

Definitions

  • the present invention relates to a registration and contour correction circuit and method for a solid-state camera. More particularly, the present invention relates to a registration and contour correction circuit and method that can electrically perform the registration correction of a solid-state image pickup device, as well as adaptively correct pixel by pixel the deterioration of spatial frequency components involved in the electrical registration correction.
  • registration errors of a solid-state image pickup device can arise if the relative positions of two or more images among red, green and blue images or portions thereof do not match each other.
  • Such a registration error of a solid-state image pickup device arises from a mounting error of pickup elements of red, green and blue, and chromatic aberration of an optical lens system. Although the mounting of pickup elements is performed by using a very precise jig, perfect obviation of the registration error is almost impossible.
  • a second object of the present invention is to provide a contour correction circuit that can reduce blurs resulting from the registration correction, thereby improving the quality of images.
  • the registration correction is carried out by interpolation calculations between the neighboring pixels by using different coefficients for respective pixels to coincide the images having registration errors.
  • a third object of the present invention is to provide a contour correction circuit that can adaptively correct, pixel by pixel, contours of an image on real time basis according to the interpolation coefficients for respective pixels.
  • a registration correction circuit for a solid-state camera having a plurality of solid-state image pickup elements each of which produces a color video signal of an image in the form of pixel data
  • the registration correction circuit which corrects registration errors between the image pickup elements, comprising: first storing means for storing a plurality of color video signals produced from the solid-state image pickup elements in the form of pixel data; second storing means for storing registration correction data predetermined corresponding to the registration errors of the respective image pickup elements; generating means for producing address data of the first storing means and interpolation coefficients according to the registration correction data in synchronism with one of the color video signals selected as a reference signal; coordinate transformation means for moving pixels by a multiple of 1-pixel pitch in real time by reading from or writing to the first storing means the pixel data in accordance with the address data; and interpolation means for performing interpolation calculations on the pixel data by using the interpolation coefficients, and for moving the pixel data by the amount less than 1-
  • the interpolation means perform infield interpolation for motion images and inframe interpolations for still images.
  • the coordinate transformation means can prepare a plurality sets of correction waveform data in advance, and to selectively use one set of the correction waveform data to handle the change of an optical lens system, or the change in chromatic aberration due to the change in zoom ratios of a zoom lens.
  • a contour correction circuit for a solid-state camera having a plurality of solid-state image pickup elements each of which produces a color video signal in the form of pixel data
  • the contour correction circuit comprising: first storing means for storing a plurality of color video signals produced from the solid-state image pickup elements in the form of pixel data; second storing means for storing registration correction data predetermined corresponding to the registration errors of the respective image pickup elements; generating means for producing address data of the first storing means and interpolation coefficients according to the registration correction data in synchronism with one of the color video signals selected as a reference signal; coordinate transformation means for moving pixels by a multiple of 1-pixel pitch in real time by reading from or writing to the first storing means the pixel data in accordance with the address data; interpolation means for performing interpolation calculations on the pixel data by using the interpolation coefficients, and for moving the pixel data by the amount less than 1-pixel pitch; means for generating contour correction signals from the reference signal in accord
  • a registration correction method for a solid-state camera having a plurality of solid-state image pickup elements each of which produces a color video signal of an image in the form of pixel data
  • the registration correction method comprising the steps of: prestoring registration correction data predetermined corresponding to the registration errors of the image pickup elements; storing a plurality of color video signals produced from the solid-state image pickup elements in the form of pixel data; generating address data and interpolation coefficients according to the registration correction data in synchronism with one of the color video signals selected as a reference signal; moving the pixel data by a multiple of 1-pixel pitch in real time by changing reading addresses or writing addresses of the pixel data in accordance with the address data; and performing interpolation calculations on the pixel data by using the interpolation coefficients so as to move the pixel data less than the 1-pixel pitch, wherein the pixel data of the color video signals other than the reference signal are moved so that the pixel data of all the color video signals will agree in their positions.
  • a contour correction method for a solid-state camera having a plurality of solid-state image pickup elements each of which produces a color video signal in the form of pixel data
  • the contour correction method comprising the steps of: prestoring registration correction data predetermined corresponding to the registration errors of the image pickup elements; storing a plurality of color video signals produced from the solid-state image pickup elements in the form of pixel data; generating address data and interpolation coefficients according to the registration correction data in synchronism with one of the color video signals selected as a reference signal; moving the pixel data by a multiple of 1-pixel pitch in real time by changing reading addresses or writing addresses of the pixel data in accordance with the address data; performing interpolation calculations on the pixel data by using the interpolation coefficients so as to move the pixel data less than the 1-pixel pitch, generating contour correction signals from the reference signal in accordance with the ratios of the interpolation coefficients; and adding the contour correction signals to the color video signals other than the reference signal,
  • the positional relationship of an optical lens system of the camera and each image signal produced from the image pickup elements in the form of pixel data may differ each other owing to the structural problem or the like. This is called registration errors.
  • the present invention corrects the registration errors as follows: first one image signal is selected as a reference signal; second, the other image signals on a random access memory or the like are shifted by using coordinate transformation and interpolation between the pixels so that the positional relationship between the optical lens system and the image signals other than the reference signal will agree to that of the reference signal.
  • the address transformation of the random access memory and the interpolation calculations to shift the pixels can be performed by using correction data prepared for each pixel and prestored in a memory.
  • enormous capacity of memory is required because correction information must be stored for each pixels.
  • the present invention digitally generates waveforms such as sawtooth waveforms and parabolic waveforms as in conventional analog image pickup tubes which use these waveforms to control the deflection systems of the tubes, and then controls the mixing ratios of respective waveforms independently by the predetermined and prestored values.
  • the correction waveform data can be freely controlled by multiplying the waveforms such as DC, SAWTOOTH, PARABOLIC by the correction coefficients for controlling the amplitudes of these waveforms.
  • the infield interpolation and the inframe interpolation can be selectively used: the infield interpolation is preferable for the motion images; and the inframe interpolation is preferable for the still images.
  • a plurality sets of coefficients prepared in advance can be selectively used to handle the change of an optical lens system, or the change of chromatic aberration due to the change in zoom ratios of a zoom lens.
  • the interpolation coefficients used in the registration correction of the present invention are supplied on a feedforward basis from a coefficient generator or the like. This means that the portions of an image to be subjected to the interpolation, and the manner in which the interpolation is performed are determined in advance. This further means that the location and degree of the deterioration of the spatial frequency characteristics are determined beforehand.
  • every part of an object picked up is supposed to take exactly the same positions in the interpolated video signals and in the reference signal because the pixels of the interpolated signals are shifted so that they will agree with their counterparts of the reference signal by the registration correction means.
  • the interpolation coefficients and for the interpolation are converted into correction coefficients that make it possible to extract the frequency components supposed to be lost by the interpolation when the reference signal is multiplied by the correction coefficients; and the frequency components thus extracted are used as correction signals to compensate the degradation in the frequency characteristics due to the interpolation.
  • Similar processing can be carried out in an interlaced scanning system in which the interpolation calculations can be switched between the inframe and infield modes in response to the detection signal of a motion detector.
  • the detection signal is simultaneously supplied to a circuit for producing the correction signal and a circuit for correcting the video signals by using the correction signal.
  • FIGS. 1A, 1B, 2A, 2B, 3A, 3B, 4A-4D and 5A to 5H are schematic diagrams illustrating a conventional registration error correction method performed in an image pickup tube
  • FIG. 6 is a schematic diagram illustrating an example of a registration error of a color image pickup device
  • FIGS. 7A-7C, 8A-8B, and 10A-10B are diagrams illustrating interpolation calculations using 4-neighboring pixels according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram illustrating 4-neighboring pixels when the interlaced scanning is carried out in the embodiment of the present invention.
  • FIGS. 11 and 12 are schematic diagrams for explaining the degradation of spatial frequency characteristics owing to the interpolation of the embodiment
  • FIGS. 13A, 13B, 14 and 15A-15D are schematic diagrams illustrating the principle of the contour enhancement of the embodiment
  • FIG. 16A and 16B are block diagrams showing the entire arrangement of the embodiment of the present invention.
  • FIG. 17 is a block diagram showing a detailed arrangement of a coefficient generator of FIG. 16B.
  • the present invention is carried out on the assumption that a solid-state image pickup device is used. To simplify the explanation, however, general description of the registration will be given with reference to various deflection waveforms applied to a conventional image pickup tube.
  • FIG. 1A illustrates a deflection waveform of an image pickup tube.
  • the deflection waveform has a sawtooth waveform. Its amplitude h corresponds to the rightmost edge of a screen in the case of horizontal deflection, whereas it corresponds to the bottom of the screen in the case of vertical deflection.
  • a constant value (form example, a 12-bit digital value) is generated.
  • the integer portion of the constant (for example, upper 4 bits) is used to control the horizontal address of the memory, and the fraction portion (for example, lower 8 bits) is used to control an interpolation coefficient which will be described later.
  • a sawtooth waveform which is specified, for example, by a 12-bit digital value, and the zero point of which is placed at the center of the screen is generated.
  • the integer portion of the digital value (the upper 4 bits) is added to the read address of the memory to produce a new read address, and the fraction portion (the lower 8 bit) is used as an interpolation coefficient to more accurately control the size of the image.
  • the above-mentioned linearity technique can be applied to the registration correction.
  • a parabolic waveform which is specified, for example, by a 12-bit digital value is generated.
  • the integer portion of the digital value (the upper 4 bits) is added to the read address of the memory to produce a new read address, and the fraction portion (the lower 8 bit) is used as an interpolation coefficient to more accurately control the linearity.
  • These waveforms are independently generated for the horizontal axis and the vertical axis, and are multiplied by the correction coefficients generated by a microcomputer, and the resultant products are added to produce horizontal correction waveform data and vertical correction waveform data.
  • the correction coefficients can be controlled by using variable resistors and rotary encoders connected to the microcomputer: the variable resistors are manipulated to adjust the registration; and the rotary encoders inform the microcomputer of the values of the variable resistors.
  • the correction waveform data thus obtained are used to control the registration correction: the integer portion (the upper 4 bits) of the data is used to control the read address of the image data, and the fraction portion (the lower 8 bits) is used to control the interpolation coefficients.
  • a red image R-ch does not agree with the green image G-ch which is specified as the reference image; a pixel (5, d) of the green image does not correspond to the pixel (5', d'), but corresponds to four pixels of the red image, (3', b'), (4', b'), (3', c') and (4', c').
  • This misregistration can be corrected by reading out the information of these addresses (3', b'), (4', b'), (3', c') and (4', c') of R-ch, and performing interpolation to these four pixels by multiplying the pixels by coefficients corresponding to the amount of registration errors.
  • misregistration occurs if the pixel (5', d') of the R-ch is read out with the corresponding pixel (5, d) of the G-ch.
  • interpolation calculations are performed by using four neighboring pixels.
  • the above method can be extended to two dimensional registration shown in FIG. 8A.
  • the horizontal registration error is m
  • the vertical registration error is n
  • the interpolation coefficients a, b, c, d of the 4-neighboring pixels take values shown in FIG. 8B.
  • the neighboring pixels in the vertical direction belongs to different fields as shown in FIG. 9.
  • pixels PA, PB, PC and PD belong to a first field
  • pixels PC' and PD' belong to a second field.
  • the interpolation coefficients in this case take values shown in FIG. 10A: interpolation coefficients a, b, c and d in FIG.
  • misregistration the amount of which is an integer multiplication of a pixel pitch
  • the interpolation calculations are not required, and so the degradation of the frequency characteristics of the image does not occur.
  • FIG. 11 illustrates the degradation of the frequency characteristics.
  • the degradation of the frequency characteristics exhibits a roll-off characteristic the maximum attenuation of which occurs at the spatial frequency of 1/(2 ⁇ ), where ⁇ is one pixel pitch.
  • the different interpolation coefficients will cause shading-like color irregularity in the image as shown in FIG. 11 which has a single frequency component over the entire image.
  • the interpolation calculations are also performed in the vertical direction.
  • the color irregularity becomes a three dimensional phenomenon. The correction of such a phenomenon has not been achieved.
  • the present invention is provided with a contour enhancement function for correcting the degradation of the spatial frequency characteristic of images.
  • the principle of the contour correction will be described.
  • FIG. 7A let us assume that an impulse-like pixel having "1" level in an image is shifted in the right-hand direction by m (0 ⁇ m ⁇ 0.5) pixel pitch by using the linear interpolation.
  • the interpolation calculations are performed for each pixel of the 4-neighboring pixels by using the interpolation coefficients of FIG. 7B, resulting in the pixels of levels shown in FIG. 7C.
  • a reverse characteristic shown in FIG. 13B is used: the coefficients corresponding to the characteristic of FIG. 13B are shown in FIG. 14. These coefficients can be used to extract frequency components having the characteristic shown in FIG. 13B from an image.
  • the G-ch signal of FIG. 6 is selected as a reference signal, and is multiplied by these coefficients.
  • the above-mentioned frequency component is extracted from the reference signal, because the registration errors between the interpolated signal (for example, R-ch signal of FIG. 6) and the reference signal are supposed to be canceled out by the interpolation, and so the interpolated pixels are assumed to agree to their counterparts of the reference signal.
  • the extracted frequency components are added to the interpolated signal in order to eliminate blurs by the interpolation. This makes it possible to remarkably enhance a contour, especially that of the luminance signal.
  • the contour correction can be extended to a two-dimensional image.
  • the interpolation coefficients are given by FIGS. 8A and 8B.
  • coefficients to be multiplied by the reference signal to correct blurs produced by the interpolation are given by FIGS. 15A-15D. These coefficients A-D are multiplied by the 4-neighboring pixels of the reference signal.
  • the contour correction is an interlaced scanning system is performed as follows.
  • vertical neighboring pixels belong to different fields as shown in FIG. 9. Consequently, vertical neighboring pixels must be switched depending on an infield or inframe mode.
  • the infield mode which uses correction coefficients A, B, C and D in FIG. 10B corresponding to the interpolation coefficients a, b, c and d in FIG. 10A, respectively, is used for a motion image, that is, when the image correlation between fields is small.
  • the inframe mode which uses correction coefficients A', B', C', and D' in FIG. 10B corresponding to interpolation coefficients a', b', c' and d' in FIG. 10A, respectively, is used for a still image, that is, when the correlation between fields is great.
  • the movement of images is detected from the reference signal, and the two mode is switched in response to the detected result.
  • the correction coefficients A, B, C and D, and the correction coefficients A', B', C' and D' are added at appropriate ratios so that transitions from the motion to still images are smoothed.
  • the frequency degradation resulting from the interpolation can be compensated by the contour correction using the correction coefficients that are adaptively determined in accordance with the interpolation coefficients associated with respective portions of the image. It is obvious that the frequency characteristics of pixels not subjected to the interpolation are not degraded.
  • FIG. 16A and 16B are block diagrams showing the entire arrangement of an embodiment of a registration and contour correction circuit for a solid-state camera according to the present invention.
  • This embodiment corrects the registration errors of a 3-original-color interlaced-scanning type solid-state image pickup device by electrically adjusting the red and blue channels so as to agree with the green channel selected as the reference signal.
  • the embodiment enhances contours of an image by compensating the spatial frequency characteristics degraded by the interpolation for correcting the registration errors.
  • the circuit is divided into six large blocks: an R-channel registration signal generating circuit 100, a B-channel registration signal generating circuit 200, a reference signal processing circuit 300, an R-channel contour correction signal generating circuit 400, a B-channel contour correction signal generating circuit 500, and an interpolation coefficient control circuit 600.
  • THe R-channel and B-channel registration signal generating circuits 100 and 200 have the same arrangement, and the R-channel and B-channel contour correction signal generating circuits 400 and 500 have the same arrangement.
  • the R-channel registration signal generating circuit 100 has a memory portion and an interpolation calculation portion.
  • the memory portion is composed of two 1-field delay circuits 1 and 2, two 1-line delay circuits 3 and 3', three random access memories (RAMs) 4, 5 and 6, and a multiplexer 7.
  • the interpolation calculation portion is composed of two 1-pixel delay circuits 8 and 9, four coefficient circuits (multipliers) 10, 11, 12 and 13, and an adder 14.
  • the R-channel contour correction signal generating circuit 400 is composed of two 1-pixel delay circuits 22 and 23, coefficient converter 24, four coefficient circuits (multipliers) 25, 26, 27 and 28, and an adder 29.
  • the interpolation coefficient control circuit 600 is composed of coefficient generator 32, synchronizing signal generator 33, microcomputer 34, and registration error controller 35.
  • the registration errors are corrected by shifting the R-channel and B-channel signals so that these signals coincide with G-channel signal specified as the reference signal.
  • the registration errors integer multiple of 1-pixel pitch can be corrected by a coordinate transformation using the RAMs 4, 5, and 6, and the registration errors less than 1-pixel pitch can be corrected by the interpolation using 4-neighboring pixels as described before.
  • the R-channel (B-channel) input signal is transmitted through the two 1-field delay circuits 1 and 2, one 1-line delay circuit 3.
  • the input signal is delayed by 1 frame +1 line, and this signal (signal [1] of FIG. 16A) is used as a reference of timing.
  • a signal of the next field (signal [3] of FIG. 16A), and a signal of next line (signal [2] of FIG. 16A) are also produced.
  • the multiplexer 7 when the input image is a motion image, the multiplexer 7 operates in the infield mode, and produces signals of the adjacent lines of the same field, i.e., signals [1] and [2], as the output signals [4] and [5]; when the input image is a still image, the multiplexer 7 operates in the inframe mode, and produces signals of adjacent lines of the same frame, i.e., signals [1] and [3] (or signals [3] and [2]), as the output signals [ 4] and [5] as shown in FIG. 16A (see FIG. 9 also).
  • the RAMs 4, 5 and 6 perform the coordinate transformation by having addresses thereof controlled by the integer portion of the coefficient supplied from the coefficient generator 32, thereby to correct the registration errors the amount of which is a multiple of 1-pixel pitch.
  • a registration error less than a pixel pitch can be corrected by the above-described interpolation calculations between 4-neighboring pixels carried out by the registration signal generating circuit 100.
  • the coefficient circuits 10-13 use the fraction portion of the interpolation coefficients fed from the coefficient generator 32.
  • FIG. 17 shows the detailed arrangement of the coefficient generator 32 together with the circuits connected thereto.
  • the coefficient generator 32 comprises a horizontal registration error control system and a vertical registration error control system.
  • the horizontal registration error control system includes a plurality of waveform generators HG1-HGi for generating predetermined correction signals, multipliers HM1-HMi for controlling the correction amounts, and an adder HA for summing up the outpust of the multipliers HM1-HMi.
  • the vertical registration error control system includes a plurality of waveform generators VG1-VGj for generating predetermined correction signals, multipliers VM1-VMj for controlling the correction amounts, and an adder VA for summing up the outputs of the multipliers VM1-VMj.
  • waveforms are digitally generated, and the upper n bits of the waveforms define the integer portion, while the lower m bits of which define the fraction portion.
  • the waveforms thus generated are fed to the multipliers HM1-HMi, and VG1-VGj, respectively, so that their mixing ratios are controlled by signals from the microcomputer 34.
  • the outputs of the multipliers HM1-HMi, and VM1-VMj are summed up by the adders HA and VA, respectively.
  • the data specifying the mixing ratios of the waveforms produced from the waveform generators are predetermined by actually picking up a test chart, manually adjusting the images by controlling the variable resistors of the registration error controller 35 so that the registration errors are canceled, calculating, by the microcomputer 34, the adjusting amounts of the variable resistors on the basis of the outputs from the rotary encoders coupled to the respective raviable resistors, and storing the adjusting results in a memory of the microcomputer 34.
  • the summed up data produced from the horizontal adder HA and the vertical adder VA are fed to a coordinate transformation control portion CTCP for calculating the address of the RAMS 4-6, and an interpolation control portion ICP for calculating the interpolation coefficients.
  • the coordinate transformation control portion CTCP produces address data of the RAMs 4-6 by adding the integer portion of the summed up data and the horizontal and vertical address reference data from a reference address signal generator RAG, while the interpolation control portion ICP produces the four interpolation coefficients for 4-neighboring pixels by using the fraction portion of the summed up data.
  • the addresses and the interpolation coefficients are switched between those for infieldmode pixels and inframe-mode pixels according to the motion signal as described above.
  • Each of the waveform generators HG1-HGi (or VG1-VGj) generates a different waveform such as follows.
  • the registration errors are corrected: the registration errors the amount of which is a multiple of 1-pixel pitch can be corrected by using the coordinate transformation on the RAMs 4-6 which is performed on the R-channel and B-channel signals with reference to the G-channel signal specified as the reference signal; and the registration errors the amount of which includes a component less than a pixel can be corrected by the interpolation calculations using the 4-neighboring pixels.
  • the registration errors of a multiple of 1-pixel pitch which is performed by the coordinate transformation produces no effect on the frequency characteristics of the video signals
  • the registration errors including errors less than a pixel, such as that of 0.5 pixel or 1.3 pixels exhibits frequency degradation because of the interpolation calculations.
  • the frequency degradation can be compensated as follows.
  • the G-channel signal specified as the reference signal is handled in a manner similar to the R-channel and G-channel signals.
  • the input G-channel signal is transmitted through the two 1-field delay circuits 15 and 16, one 1-line delay circuit 17.
  • a reference signal delayed by 1 frame +1line than the input signal is produced.
  • a signal delayed by 1-field than the reference signal and a signal delayed by 1-line than the reference signal are produced from the 1-line delay circuit 17' and the 1-field delay circuit 16, respectively.
  • the signal of 1-field delay and the signal of 1-line delay are switched by the multiplexer 21 in response to the motion signal generated from the motion detector 31: when the motion signal indicates a still image, the signal of 1-field delay is produced from the multiplexer 21, whereas when the motion signal indicates a motion image, the signal of 1-line delay is produced from the multiplexer 21.
  • the signal thus obtained and the signal of 1-field and 1-line delay is transmitted through the 1-pixel delay circuits 22 and 23, respectively.
  • the 4-neighboring pixels are obtained (see FIG. 9).
  • the G-channel signal must be delayed by n-lines with regard to the R-channel and B-channel signals.
  • the n-line delay circuits 18-20 are used for this purpose.
  • the 4-neighboring pixels thus obtained are fed to the coefficient circuits 25-28.
  • the coefficient circuits 25-28 multiply the 4-neighboring pixels by the correction coefficients A-D or A'-D' supplied from the coefficient converter 24, respectively.
  • the coefficient converter 24 converts the fraction portions of the interpolation coefficients a-d or a'-d' fed from the coefficient generator 32 into the correction coefficients A-D, or A'-D', which exhibit the inverse characteristics of those of the interpolation coefficients a-d, or a'-d', and supplies them to the coefficients circuits 25-28 in real time.
  • the outputs of the coefficient circuits 25-28 are summed up by the adder 29.
  • FIGS. 15A-15D The relationship between the interpolation coefficients a-d or a'-d' and the correction coefficients A-D or A'-D' is shown in FIGS. 15A-15D, and each of these coefficients corresponds to one of the 4-neighboring pixels.
  • the correction signal produced from the adder 29 is supplied to an adder 30 which adds the correction signal to the R-channel signal.
  • An adder 30' performs similar addition with regard to the B-channel signal.
  • each of the multiplexers 7 and 21 can be replaced by an adaptive adding circuit that mixes the calculation results of the infield and inframe calculation results at particular ratios. This makes it possible to smooth transitions from motion to still images or from still to motion images.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Color Television Image Signal Generators (AREA)
US07/810,360 1989-11-09 1991-12-18 Registration and contour correction circuit and method for solid-state camera Expired - Lifetime US5157481A (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP1-290098 1989-11-09
JP1-290099 1989-11-09
JP1290098A JP2552741B2 (ja) 1989-11-09 1989-11-09 レジストレーション補正回路
JP1290099A JP2552742B2 (ja) 1989-11-09 1989-11-09 適応型輪郭補正回路

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US07604012 Continuation 1990-10-26

Publications (1)

Publication Number Publication Date
US5157481A true US5157481A (en) 1992-10-20

Family

ID=26557883

Family Applications (1)

Application Number Title Priority Date Filing Date
US07/810,360 Expired - Lifetime US5157481A (en) 1989-11-09 1991-12-18 Registration and contour correction circuit and method for solid-state camera

Country Status (3)

Country Link
US (1) US5157481A (de)
EP (1) EP0427436B1 (de)
DE (1) DE69029507T2 (de)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5276512A (en) * 1991-03-07 1994-01-04 Matsushita Electric Industrial Co., Ltd. Video signal motion detecting method and noise reducer utilizing the motion
US5353056A (en) * 1992-10-27 1994-10-04 Panasonic Technologies, Inc. System and method for modifying aberration and registration of images
US5442394A (en) * 1991-02-26 1995-08-15 Samsung Electronics Co., Ltd. Method for improving the resolution of three CCD camera and apparatus thereof
US5457548A (en) * 1990-05-17 1995-10-10 Canon Kabushiki Kaisha Color image processing apparatus
US5973755A (en) * 1997-04-04 1999-10-26 Microsoft Corporation Video encoder and decoder using bilinear motion compensation and lapped orthogonal transforms
US6002434A (en) * 1997-01-14 1999-12-14 Matsushita Electric Industrial Co., Ltd. Registration correction waveform determination method and system for a television camera
US6061477A (en) * 1996-04-18 2000-05-09 Sarnoff Corporation Quality image warper
US20010030697A1 (en) * 1997-05-16 2001-10-18 Dischert Lee Robert Imager registration error and chromatic aberration measurement system for a video camera
US6356304B1 (en) * 1996-04-26 2002-03-12 Sony Corporation Method for processing video signal and apparatus for processing video signal
US20040013319A1 (en) * 2002-07-19 2004-01-22 Wenstrand John S. Resolution and image quality improvements for small image sensors
US20040207758A1 (en) * 2002-12-25 2004-10-21 Hitachi, Ltd. Video signal processing apparatus
US20090207271A1 (en) * 2005-03-04 2009-08-20 Nikon Corporation Image Processor Correcting Color Misregistration, Image Processing Program, Image Processing Method, and Electronic Camera
CN101271248B (zh) * 2007-03-23 2011-09-07 奥林巴斯映像株式会社 电子图像拾取装置
US20130113876A1 (en) * 2010-09-29 2013-05-09 Huawei Device Co., Ltd. Method and Device for Multi-Camera Image Correction

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0518185B1 (de) * 1991-06-10 1996-12-11 Eastman Kodak Company Kreuzkorrelationsausrichtsystem für einen Bildsensor
US5475428A (en) * 1993-09-09 1995-12-12 Eastman Kodak Company Method for processing color image records subject to misregistration

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0059076A1 (de) * 1981-02-19 1982-09-01 Tektronix, Inc. Konvergenzregelsystem für eine Mehrstrahlkathodenstrahlröhre
EP0084195A1 (de) * 1981-12-30 1983-07-27 Koninklijke Philips Electronics N.V. Verfahren zur automatischen Einstellung einer Fernsehkamera und dafür geeignete Fernsehkamera, geeignete Speicher und Linsensystem, und geeignete Speicher
GB2135851A (en) * 1983-02-25 1984-09-05 Rca Corp Solid state tv camera image control
EP0176240A1 (de) * 1984-08-31 1986-04-02 Rca Licensing Corporation Digitale Videoverzögerung mit Abtastwertinterpolation
JPS6189791A (ja) * 1984-10-08 1986-05-07 Sony Corp 固体カラ−撮像装置
US4835594A (en) * 1987-04-22 1989-05-30 Sony Corporation Registration adjusting apparatus for a color television camera using a plurality of pick-up tubes
EP0350794A2 (de) * 1988-07-13 1990-01-17 Hitachi, Ltd. Konvergenzfehler-Korrektur für Festkörperbildaufnahmegeräte

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0059076A1 (de) * 1981-02-19 1982-09-01 Tektronix, Inc. Konvergenzregelsystem für eine Mehrstrahlkathodenstrahlröhre
EP0084195A1 (de) * 1981-12-30 1983-07-27 Koninklijke Philips Electronics N.V. Verfahren zur automatischen Einstellung einer Fernsehkamera und dafür geeignete Fernsehkamera, geeignete Speicher und Linsensystem, und geeignete Speicher
GB2135851A (en) * 1983-02-25 1984-09-05 Rca Corp Solid state tv camera image control
EP0176240A1 (de) * 1984-08-31 1986-04-02 Rca Licensing Corporation Digitale Videoverzögerung mit Abtastwertinterpolation
JPS6189791A (ja) * 1984-10-08 1986-05-07 Sony Corp 固体カラ−撮像装置
US4835594A (en) * 1987-04-22 1989-05-30 Sony Corporation Registration adjusting apparatus for a color television camera using a plurality of pick-up tubes
EP0350794A2 (de) * 1988-07-13 1990-01-17 Hitachi, Ltd. Konvergenzfehler-Korrektur für Festkörperbildaufnahmegeräte

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
International Broadcasting Convention, IBC 80, Sep. 20 23, 1989, pp. 27 33, ii xviii. *
International Broadcasting Convention, IBC 80, Sep. 20-23, 1989, pp. 27-33, ii-xviii.

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5457548A (en) * 1990-05-17 1995-10-10 Canon Kabushiki Kaisha Color image processing apparatus
US5442394A (en) * 1991-02-26 1995-08-15 Samsung Electronics Co., Ltd. Method for improving the resolution of three CCD camera and apparatus thereof
US5276512A (en) * 1991-03-07 1994-01-04 Matsushita Electric Industrial Co., Ltd. Video signal motion detecting method and noise reducer utilizing the motion
US5353056A (en) * 1992-10-27 1994-10-04 Panasonic Technologies, Inc. System and method for modifying aberration and registration of images
US6061477A (en) * 1996-04-18 2000-05-09 Sarnoff Corporation Quality image warper
US6356304B1 (en) * 1996-04-26 2002-03-12 Sony Corporation Method for processing video signal and apparatus for processing video signal
US6002434A (en) * 1997-01-14 1999-12-14 Matsushita Electric Industrial Co., Ltd. Registration correction waveform determination method and system for a television camera
US6307887B1 (en) 1997-04-04 2001-10-23 Microsoft Corporation Video encoder and decoder using bilinear motion compensation and lapped orthogonal transforms
US5973755A (en) * 1997-04-04 1999-10-26 Microsoft Corporation Video encoder and decoder using bilinear motion compensation and lapped orthogonal transforms
US20010030697A1 (en) * 1997-05-16 2001-10-18 Dischert Lee Robert Imager registration error and chromatic aberration measurement system for a video camera
US20040013319A1 (en) * 2002-07-19 2004-01-22 Wenstrand John S. Resolution and image quality improvements for small image sensors
US6983080B2 (en) 2002-07-19 2006-01-03 Agilent Technologies, Inc. Resolution and image quality improvements for small image sensors
US20040207758A1 (en) * 2002-12-25 2004-10-21 Hitachi, Ltd. Video signal processing apparatus
US7304672B2 (en) * 2002-12-25 2007-12-04 Hitachi, Ltd. Contour correcting video signal processing apparatus
US20090207271A1 (en) * 2005-03-04 2009-08-20 Nikon Corporation Image Processor Correcting Color Misregistration, Image Processing Program, Image Processing Method, and Electronic Camera
US7945091B2 (en) 2005-03-04 2011-05-17 Nikon Corporation Image processor correcting color misregistration, image processing program, image processing method, and electronic camera
CN101271248B (zh) * 2007-03-23 2011-09-07 奥林巴斯映像株式会社 电子图像拾取装置
US20130113876A1 (en) * 2010-09-29 2013-05-09 Huawei Device Co., Ltd. Method and Device for Multi-Camera Image Correction
US9172871B2 (en) * 2010-09-29 2015-10-27 Huawei Device Co., Ltd. Method and device for multi-camera image correction

Also Published As

Publication number Publication date
EP0427436A3 (en) 1992-01-02
DE69029507T2 (de) 1997-05-15
DE69029507D1 (de) 1997-02-06
EP0427436A2 (de) 1991-05-15
EP0427436B1 (de) 1996-12-27

Similar Documents

Publication Publication Date Title
US5157481A (en) Registration and contour correction circuit and method for solid-state camera
KR0153509B1 (ko) 촬상장치
US5353056A (en) System and method for modifying aberration and registration of images
US5526053A (en) Motion compensated video signal processing
US5253064A (en) Video camera apparatus with electronic zoom control and method therefor
KR0125108B1 (ko) 정지화상기록 디지탈 카메라
US5333055A (en) Video camera circuit for processing image signals from an image pickup device having a mosaic color filter
GB2231228A (en) Video signal to photographic film conversion
JP4212134B2 (ja) 画像信号処理装置
US5113247A (en) Solid state image pickup apparatus for correcting discrepancy of registration
US5793447A (en) Digital convergence apparatus
EP0313026B1 (de) Verfahren und Vorrichtung zur Bewegungsvektorkompensation im Empfang eines auf MUSE-System basierenden Fernsehsignals
KR930011968B1 (ko) 디지탈 수직 비임 랜딩 보정 회로
US4771334A (en) Digital video delay by sample interpolation
US6917383B2 (en) Image sensing device
US6380973B1 (en) Image sensing apparatus and method for image processing a moving image or a still image
JPH0224073B2 (de)
US5784502A (en) Interpolation apparatus
JP2552742B2 (ja) 適応型輪郭補正回路
JPH04275776A (ja) 画像読み取り装置
JPH0795595A (ja) カラー撮像装置
JPH05103226A (ja) 輪郭平滑化回路
JPH03151784A (ja) レジストレーション補正回路
CA1256982A (en) Digital video delay by sample interpolation
JPH041556B2 (de)

Legal Events

Date Code Title Description
STCF Information on status: patent grant

Free format text: PATENTED CASE

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Free format text: PAYER NUMBER DE-ASSIGNED (ORIGINAL EVENT CODE: RMPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FPAY Fee payment

Year of fee payment: 12